hAMSCs. Amnion samples were obtained from Shengjing Hospital of China Medical University (Shenyang, China) and all samples were obtained with patient informed consent. Freshly isolated amniotic membranes were washed with phosphate-buffered saline (PBS) to remove the blood and then immersed in an antibiotic solution for 20 min. Then, the amnions were cut into ∼25 cm² pieces, transferred into a 0.25% trypsin solution (0.25% trypsin; Gibco) and incubated at 37°C for 1 h. The digested tissue was centrifuged, and the supernatant was discarded; this process was repeated three times. Culture medium was added after the last digestion to neutralize the digestive enzymes and end the digestion. The amniotic membrane pieces were then incubated at 37°C in PBS containing 1 mg/ml collagenase (Sigma-Aldrich, St. Louis, MO, USA) and 0.1 mg/ml DNase (Takara Bio, Inc., Japan) for 1 h. An equal amount of culture medium was then added to end digestion. Following this digestion, the resulting cell suspension was filtered through a cell sieve to collect the cells, which were plated into a culture flask (designated as P₀) in Dulbecco’s modified Eagle’s medium (DMEM; Gibco) containing 10% fetal bovine serum (FBS; Hyclone), 100 U/ml penicillin and 100 μg/ml streptomycin. The flasks were transferred into a 37°C, 5% CO₂ saturated humidity incubator to continue culturing. The culture medium was changed every two or three days. The cells were passaged at a 1:3 ratio when they reached 80–90% confluency. For the experiment, hAMSCs were used at passages 3–6.

hBMSCs. Bone marrow samples were obtained from Shengjing Hospital of China Medical University (Shenyang, China) and all samples were obtained with patient informed consent. Approximately 10 to 30 ml of bone marrow aspirate was collected in a syringe containing 10,000 IU heparin to prevent coagulation. Bone marrow aspirates were mixed with DMEM containing 10% FBS and centrifuged at 1,200 × g for 5 min to obtain a cell pellet. These cells were seeded into 25 cm² tissue culture flasks and cultured in DMEM containing 10% FBS, 100 U/ml penicillin, and 100 μg/ml streptomycin (4). After 48 h, non-adherent cells were removed, and fresh medium was added. The medium was changed every two or three days. The adherent spindle-shaped cells were further propagated for three passages. For the experiment, hBMSCs were used at passages 3–6.

hADSCs. Adipose samples were obtained from Shengjing Hospital of China Medical University and all samples were obtained with patient informed consent. To isolate hADSCs, adipose tissues were washed with PBS and digested with 0.075% type I collagenase under gentle agitation at 37°C for 30 min (4). Enzyme activity was neutralized with DMEM containing 10% FBS, and the suspensions were centrifuged at 300 × g for 10 min to obtain a pellet. The pellet was filtered with a 70 μm nylon mesh filter, suspended in DMEM containing 10% FBS, 100 U/ml penicillin and 100 μg/ml streptomycin. The cultures were maintained at 37°C in a humidified atmosphere containing 5% CO₂. After 48 h, the non-adherent cells were removed, and fresh medium was added to the cells. When the monolayer of adherent cells reached confluence, they were trypsinized, resuspended in DMEM containing 10% FBS. For the experiment, hADSCs were used at passages 3–6.

Culture of dermal fibroblasts. Skin samples were obtained from Shengjing Hospital of China Medical University and all samples were obtained with patient informed consent. Fresh healthy skin tissue was de-epithelialized and minced into 2 mm pieces (3). These pieces were spread on the surface of a 100 mm tissue culture plate that was precoated with 3 ml of DMEM containing 10% FBS, 100 U/ml penicillin and 100 μg/ml streptomycin. The plates were incubated at 37°C for 4 h to facilitate tissue adhesion. Following this incubation, 12 ml of DMEM containing 10% FBS was added, and the plates were returned to the incubator at 37°C with 5% CO₂. When the fibroblasts reached 80–90% confluency, the cells were passaged for expansion. For the experiments, human dermal fibroblast (hDF) cells were used at passages 3–6.

Flow cytometry was used to detect stem cell-related cell surface markers on the cultured cells. MSCs (hAMSCs, hBMSCs or hADSCs) were cultured in their corresponding growth medium for 72 h prior to analysis. The cells were trypsinized with 0.05% trypsin-EDTA, collected in regular mesenchymal media and centrifuged for 5 min at 800 × g at 4°C. Then, the cells were resuspended in PBS containing 1% bovine serum albumin at 10⁶ cells/ml and incubated with monoclonal PE-conjugated antibodies for CD45, CD34, CD14 and HLA-DR, APC-conjugated antibodies for CD45, and FITC-conjugated antibodies for CD90 and CD105 (all were from Becton-Dickinson, USA) for 45 min on ice. The cells were subsequently washed with PBS, fixed with 1% formaldehyde, and analyzed by flow cytometry. A minimum of 10,000 cell events per assay were acquired on a FACSCalibur flow cytometer.

MSCs (hAMSCs, hBMSCs or hADSCs) between passages 3 and 6 were tested for their ability to differentiate into osteocytes, chondrocytes and adipocytes (21). To induce osteogenic differentiation, cells were cultured for three weeks in DMEM with 10% FBS, 0.1 μM dexamethasone (Amresco), 50 μg/ml ascorbic acid and 10 mM glycerophosphate (were from Sigma-Aldrich), and the medium was changed every three days. After 21 days, the cells were analyzed by alkaline phosphatase staining. To induce adipogenic differentiation, cells were cultured for two weeks in DMEM with 10% FBS, 1 μM dexamethasone, 0.5 μM isobutylmethylxanthine, 60 μM indomethacin and 170 μM insulin (Sigma-Aldrich), and the medium was changed every two days. After 14 days, the cells were analyzed by Oil Red O staining (Sangon Biotech). To induce chondrogenic differentiation, cells were cultured for three weeks in DMEM with 10% FBS, 50 μg/ml ascorbic acid, 100 μg/ml sodium pyruvate (Sigma-Aldrich), 0.1 μM dexamethasone and 10 ng/ml transforming growth factor (TGF)-β1 (PeproTech), and the medium was changed every three days. After 21 days, the cells were analyzed by Alcian blue (Sangon Biotech) staining.

PEI-coated Fe₃O₄ nanoparticles are superparamagnetic iron oxide nanoparticles (SPIONs) (19,22–24). To track MSCs in vivo, cells were grown to confluence in cell culture dishes and on culture glass slides and treated with SPIONs at a final concentration of 20–30 μg/ml (MSC culture medium) for 24 h. To determine the presence of intracellular SPION particles, slide-cultured cells were fixed with 10% neutral-buffered formalin for 30 min, rinsed with PBS and then stained with Prussian blue. The slide was immersed in a solution consisting of equal parts of 20% HCl and 10% ferrocyanide for 20 min and then rinsed in tap water. Counterstaining with nuclear fast red was performed for 5 min.

All animal procedures were reviewed and approved by the China Medical University Review Committee. Mice (8-week-old female, body weight 20–23 g) were obtained from the Lab Animal Center of China Medical University. The animals were randomly divided into four groups (9 mice/group). Following hair removal from the dorsal surface, the mice were anesthetized with sodium pentobarbital (0.5 mg/g), and a 10-mm full-thickness excisional skin wound was made on the back of each mouse (25,26). MSCs were incubated with SPIONs for in vivo cell tracking using Prussian blue staining. Each wound received 1 million labeled cells (injected hAMSCs, hBMSCs or hADSCs as the experimental groups or injected PBS as the control) via 4 injections of 0.25×10⁶ cells in 60 μl of PBS, injected intradermally around the wound at four different injection sites, and dressed with fresh human amnion as an antimicrobial agent. All mice were housed individually.

Digital images of the wounds were captured at Days 0, 3, 7, 10 and 14 with a 10-megapixel digital camera (Olympus), and an image analyzer (Image-Pro Plus 6.0) was used to take wound-size measurements. Wound healing was quantified by calculating the remaining wound area for each group at 0, 3, 7, 10 and 14 days after grafting. The percentage of wound closure was calculated as follows: (area of original wound - area of actual wound)/area of original wound ×100 (15). Mice were sacrificed 7, 14 and 21 days after grafting for histological assessment. The harvested wound areas, including a border of normal tissue, were immediately fixed in 10% neutral-buffered formalin. The specimens were embedded in paraffin, sectioned, and stained with hematoxylin and eosin (H&E) to determine the quality of wound healing as measured by the characteristics and amount of granulation of the epidermal and dermal layers. To find migrated SPION-labeled MSCs in the wound, the sectioned tissue was stained with Prussian blue.

hDFs were cocultured with MSCs in transwells with 0.4 μm pores (Corning-Costar, New York, NY, USA), which allowed for paracrine interactions without direct cell contact. For all studies, hDFs were seeded at 0.5×10⁴ cells/cm² and maintained in DMEM containing 10% FBS for 3 days prior to coculture (10). MSCs were seeded at 0.5×10⁴ cells/cm² in the inserts and maintained in DMEM containing 10% FBS for 3 days prior to coculture, which allowed the MSCs to reach 70–80% confluence. MSC-containing inserts were then added to the dermal fibroblast wells and flooded with fresh medium containing 2% FBS. Controls consisted of dermal fibroblasts cocultured with an empty insert.

To verify the effect of MSCs on hDF proliferation, hDFs were cocultured with MSCs in transwells with 0.4 μm pores (Corning-Costar). The hDFs were seeded at 0.5×10⁴ cells/cm² in 24-well plates and maintained in DMEM with 10% FBS for 3 days prior to coculture. MSCs were seeded at 0.5×10⁴ cells/cm² in the inserts and maintained in the same medium for 3 days prior to coculture, which allowed the MSCs to reach 70–80% confluence (10,15). MSC-containing inserts were then added to the hDF wells and flooded with fresh media containing 2% FBS. Controls consisted of hDFs cocultured with an empty insert. At 24, 48 and 72 h, hDF proliferation was measured using a tetrazolium reagent, 2-(4-indophenyl)-3-(4-nitrophenyl)-5-(2,4-disulphophenyl)-2H-tetrazolium monosodium salt (WST-1, Cell Counting kit; Beyotime, Beijing, China). hDFs were incubated with WST-1 for 4 h at 37°C (27–30). The staining intensity in the medium was measured by determining the absorbance at 450 nm, and the data were expressed as ratios of the control value. The data are from three independent experiments performed in duplicate.

We analyzed the effect of MSCs on dermal fibroblast migration using a scratch wound assay. MSCs were seeded at 0.5×10⁴ cells/cm² in the inserts (0.4 μm pores) and maintained in the same medium for 3 days prior to coculture, which allowed the MSCs to reach 70–80% confluence (10,15). hDFs were seeded at 0.5×10⁴ cells/cm² in 6-well plates and maintained in DMEM containing 10% FBS and maintained until reaching 80–90% confluence (Day 3 post-seeding). Confluent hDFs were then scored with a sterile pipette tip to leave a scratch of ∼0.4–0.5 mm in width. Each well was washed twice with PBS to remove the cell debris and then either an empty control insert or an MSC-containing insert was placed in the scratched hDF well with DMEM containing 2% FBS. Cell migration into the scratch was photographed at 0, 24, 48 and 72 h using an inverted microscope. The results are presented as the percentage of wound healing, which was calculated as follows: [Wound Area (initial) − Wound Area−(final)]/Wound area (initial) ×100 (10,15). The data are from three independent experiments performed in duplicate.

We also measured hDF migration using a transwell migration assay (8.0 μm pore size; Corning-Costar) according to the manufacturer’s protocol. In brief, MSCs were seeded in the bottom chamber of 24-well plates at 0.5×10⁴ cells/cm²; control bottom chambers contained medium without cells. A cell culture insert with 8.0 μm pores was placed in the well, and hDF at 0.5×10⁴ cells/cm² was added to the insert and incubated at 37°C in 5% CO₂ for 12, 24 or 48 h. Following incubation, the cells were removed from the upper chamber, and the cells on the bottom side of the insert (migrated cells) were fixed with paraformaldehyde (PFA), stained with hematoxylin and photographed. Cells were counted from eight randomly selected regions per well. The data are from three independent experiments performed in duplicate.

Total RNA was extracted from hDF 48 h after the initiation of coculture with MSCs using TRIzol Reagent (Invitrogen). The concentration of extracted RNA was determined using an ultraviolet spectrophotometer. cDNA was synthesized from 500 ng of total RNA using PrimeScript™ RT reagent (Takara Bio, Inc.). The primers used for real-time PCR are shown in Table I. The reactions were performed using the SYBR PrimeScript RT-PCR kit (Takara Bio, Inc.) with an ABI 7500 Sequence Detection System (Applied Biosystems). Each experiment was performed with three biological replicates. As an internal control, levels of glyceraldehyde-3-phosphate-dehydrogenase (GAPDH) were quantified in parallel with the target genes. Normalization and fold changes were calculated using the ΔΔCt method.

The hDFs were washed three times with ice-cold PBS and prepared with RIPA buffer (50 mM Tris-HCl, 0.5% deoxycholate, 1% Nonidet P-40, 150 mM NaCl and 0.1% SDS) containing a protease inhibitor mixture (Roche Diagnostics) 48 h after the initiation of coculture with MSCs. The samples were separated by SDS-PAGE and then transferred to a polyvinylidene difluoride membrane (Millipore, Bedford, MA, USA). The membrane was blocked with 5% nonfat milk and incubated with a type I collagen primary antibody (R&D, USA) at 4°C overnight. Then the blots were incubated with an HRP-conjugated secondary Ab (Proteintech, USA) for 1 h at room temperature. Immunoreactive bands were visualized by the SuperSignal West Pico Chemiluminescent Substrate (Pierce, Rockford, IL,USA) using LAS-3000 mini (Fujifilm).

All experiments were repeated at least three times. The data are shown as the means ± SD. Comparisons between groups were analyzed using a t-test. All the statistical analyses were preformed using SPSS 19.0 computer software. P-values <0.05 were considered to indicate statistically significant differences.

MSCs were isolated from human amnion, bone marrow and adipose tissue. They were expanded on a plastic dish and exhibited similar fibroblast-like morphologies (Fig. 1A). Stem cell-related cell surface markers were assessed using flow cytometry to characterize isolated MSCs. All types of MSCs expressed CD105 and CD90, but not CD34, CD14, CD45 or HLA-DR, consistent with previous reports (15,31) (Fig. 1B). MSCs were further characterized by confirming their ability to undergo specific osteogenic, chondrogenic and adipogenic differentiation, as examined by alkaline phosphatase staining, Alcian blue staining and Oil Red O staining (Fig. 1A), respectively. Therefore, these MSC populations have the ability to differentiate into osteocytes, chondrocytes and adipocytes.

To compare the roles of hAMSCs, hBMSCs and hADSCs in wound healing (15,26), we injected them into cutaneous wounds on mice. The MSC-injected groups exhibited accelerated wound closure compared with the controls. The enhancement appeared early, at 3 days after transplantation, and became more evident after 7 days (Fig. 2A). Quantitative analysis demonstrated that an hADSC injection had the most pronounced effect on wound closure, followed by the hBMSCs, hAMSCs and the control groups at 3, 7, 10 and 14 days after transplantation (Fig. 2B). Although a slight improvement in wound healing was noted in the hAMSC and hBMSC groups, there was no statistically significant difference between the two groups. Histological evaluation of the wounds at 7, 14 and 21 days after surgery showed enhanced re-epithelialization in the hADSC group compared with all other groups (Fig. 2C). In addition, granulation tissue in the hADSC-injected group appeared to be thicker and larger. These results indicated that hADSCs had an evident effect on wound healing compared with hAMSCs or hBMSCs. To find migrated MSCs in the wound skin, we labeled MSCs with SPIONs and stained the slide-cultured cells and sectioned tissue by Prussian blue staining (Fig. 2D).

Since the proliferation of hDFs is an important aspect of wound healing (10,32), we compared the differences in the stimulatory effects of hAMSCs, hBMSCs and hADSCs on hDF proliferation in transwell coculture experiments to clarify how these cell types altered hDF proliferation. WST analysis of hDFs revealed that hDFs cultured in transwells with an MSC insert showed enhanced proliferation compared with control hDFs cultured with an empty insert after 48 and 72 h of coculture. However, there was no significant difference in fibroblast proliferation between the hAMSC, hBMSC and hADSC inserts (Fig. 3A).

To examine whether hADSCs exhibited biological effects relevant to hDF migration (10,32), we compared the effects of hAMSCs, hBMSCs and hADSCs on hDF migration. Images of scratch wounds on hDF monolayers were captured 24, 48 and 72 h after scratching to show that hDF migration into the scratch wound area was accelerated in coculture with hADSCs compared with other cocultures (Fig. 3B). Quantitative analysis confirmed that hADSCs were superior to hAMSCs and hBMSCs in promoting hDF migration (Fig. 3C).

In transwell cell migration assays, hAMSCs, hBMSCs or hADSCs were seeded in the bottom chamber of a transwell and hDFs were loaded into the top chamber. hDFs migrated into the bottom chamber and were stained 12, 24 and 48 h after initiating coculture (Fig. 3D). Quantitative analysis confirmed that hDF migration into the bottom chamber was more accelerated in the presence of hADSCs (Fig. 3E).

To understand how the molecular mechanisms of hADSCs improved wound healing, we compared the expression of wound-healing-related cytokines (25,26,32) and ECM proteins (15) in hDFs cocultured with hAMSCs, hBMSCs or hADSCs. Real-time PCR was performed for the cytokine vascular endothelial growth factor (VEGF), basic fibroblast growth factor (bFGF), keratinocyte growth factor (KGF) and TGF-β, which confirmed significantly higher mRNA expression of these cytokine genes in hDFs cocultured for 48 h with hADSCs than with other MSCs (Fig. 4A).

Next, we investigated the effects of hADSCs on hDF expression of ECM type I collagen, as measured by mRNA and protein levels. Real-time PCR analysis (Fig. 4B) showed that hADSCs have a more potent effect on type I collagen expression by hDFs than either hAMSCs or hBMSCs. Western blotting (Fig. 4C) showed that hADSCs also have a more potent effect on type I collagen expression by hDFs. These results indicate that hDFs cocultured with hADSCs for 48 h have significantly upregulated expression of type I collagen, as measured by mRNA and protein levels.